An evaluation of health warnings in cigarette advertisements using standard market research methods: what does it mean to warn?

Objective -To evaluate the effectiveness of the mandated Surgeon General's warn ing on cigarette advertisements in com parison to new warnings developed using standard advertising techniques. Design Adolescent subjects were ex posed in a controlled setting to slide images of advertisements including a Marlboro cigarette advertisement con taining either mandated or newly de veloped warnings. Setting Subjects were recruited from, and testing was completed in, high schools from the area of Augusta, Georgia (USA). Participants A convenience sample of 220 subjects ranging in age from 13 to 19 years. Intervention Controlled exposure to either a currently mandated warning or one of three newly developed warnings placed within a cigarette advertisement. Main outcome measures Post-exposure recall, masked recall, and aided recog nition tests were used to evaluate the effectiveness of the warnings' communi cation. These are standard market re search methods that are frequently em ployed to evaluate print advertisements. Results-The mandated warning per formed poorly as a communication device. It was identified as a health advisory, but failed to communicate more specific risk information. Only 15 % of subjects recalled the warning's health concept in the masked recall test. In contrast, newly developed warnings were more effective in communicating specific health information: 66% recalled the health message. The major elements of the cigarette advertisement were quickly noted and frequently recalled. The me dian cumulative exposure time required to identify the advertisement as a Marl boro cigarette advertisement was only 0.03 seconds. Conclusions The currently mandated warnings on cigarette advertisements fail to communicate specific health risk in formation effectively. Warnings which are novel, targeted, and developed through a creative process function more effectively as communicating devices. (Tobacco Control 1993; 2: 279-85) Introduction The Federal Cigarette Labeling and Adver tising Act of 1965 established the first cigarette warning label in the US. This act required that all cigarette packages be labeled with the warning "Caution: cigarette smoking may be hazardous to your health." In 1969, the Public Health Cigarette Smoking Act attempted to strengthen the warning by changing its word ing to read "Warning: the Surgeon General has determined that cigarette smoking is dangerous to your health." In 1972, the Federal Trade Commission (FTC), which is responsible for regulating cigarette advertis ing, extended this to include warning labels on all cigarette advertisements. Ongoing concerns regarding the effective ness of the warning resulted in a 1981 FTC investigation which concluded that the original warning was probably ineffective because it was: (1) overexposed and worn out; (2) no longer novel; (3) abstract and therefore diffi cult to remember; and (4) not likely to be perceived as personally relevant.1 In response to these findings, Congress passed the Com prehensive Smoking Education Act in 1984, which mandated a system of four rotating warnings for cigarette packages and advertise ments. In addition, in 1986 Congress passed the Comprehensive Smokeless Tobacco Edu cation Act, which mandated that smokeless tobacco packages and advertisements include one of three warnings printed in a " circle-and arrow" design. This design had been identified by the FTC as one that might draw more attention to the warning.1 The product liability implication of tobacco warnings has recently been clarified by the Supreme Court's ruling in the case of Thomas Cipollone v Liggett Group, Inc.2 This ruling established that the presence of the warning does not preempt all tort claims against cigarette manufacturers, but does preempt claims based on failure to include more effective warnings after 1969. Successful claims may still be brought if they are based on misrepresentations, fraud, or conspiracy; on failure-to-warn through channels other than labels and advertising; on failure-to-warn in advertising prior to 1969; or on strict liability theories. State and local regulatory authorities are subject to these same limitations. By this decision, the Supreme Court eliminated the ability of courts and regulatory agencies to modify the currently mandated warnings. University Hospital, Augusta, Georgia, USA P M Fischer College of Journalism and Mass Communication, University of Georgia, Athens, Georgia, USA D M Krugman J E Fletcher RJFox Medical College of Georgia, Augusta, Georgia, USA T H Rojas Correspondence to : Paul M Fischer, Section of Family Medicine, University Hospital, Augusta, Georgia 30912, USA This content downloaded from 157.55.39.55 on Sun, 12 Jun 2016 05:12:34 UTC All use subject to http://about.jstor.org/terms 280 Fischer, Krugman, Fletcher, Fox, Rojas Therefore, action by Congress will be required to change them. The accumulated evidence suggests that current tobacco warnings are at best limited in communicating the hazards of smoking. This finding is supported by both surveys and experimental research. In addition to the 1981 FTC report, an eye-tracking study of ado lescent subjects viewing cigarette advertise ments has shown that only 37 % of subjects look at the warning long enough to read any of its words.3 Other research has shown that cigarette warnings are too small to be read on billboard ads and taxicab ads,4,5 that they fail to communicate "risk",6 that they are not believable,7 and that they require college-level reading comprehension.8 In a study of US cigarette package warnings, the majority of smokers were unfamiliar with even the general themes of the health messages.9 An important question is, "What level of communication is necessary, or acceptable, in terms of warning consumers?" After all, communication is not all or nothing, but rather, occurs in steps and exists at various levels of persuasiveness. A great deal of communication research has focused on how information is processed. One of the most notable models divides communication into distinct stages exposure, attention, compre hension, acceptance, and retention.10 Infor mation passes through these stages prior to having an impact on either attitudes or be haviour. The objectives of the present study were to investigate, in a more comprehensive manner than previous studies, what current warnings actually communicate regarding the hazards of smoking, and to examine whether or not new warnings could provide more meaningful com munication. In so doing, the research raises the policy question: "What does it mean to warn?"